This invention relates to the production of elemental material from the halides thereof and has particular applicability to those metals and non-metals for which the reduction of the halide to the element is exothermic. Particular interest exists for titanium, and the present invention will be described with particular reference to titanium but is applicable to other metals and non-metals such as Al, As, Sb, Be, B, Ta, Ge, V, Nb, Mo, Ga, Ir, Os, U and Re, all of which produce significant heat upon reduction from the halide to the metal. For the purposes of this application, elemental materials include those metals and non-metals listed above or in Table 1.
At present, titanium production is by reduction of titanium tetrachloride, which is made by chlorinating relatively high-grade titanium dioxide ore. Ores containing rutile can be physically concentrated to produce a satisfactory chlorination feed material; other sources of titanium dioxide, such as ilmenite, titaniferous iron ores and most other titanium source materials, require chemical beneficiation.
The reduction of titanium tetrachloride to metal has been attempted using a number of reducing agents including hydrogen, carbon, sodium, calcium, aluminum and magnesium. Both the magnesium and sodium reduction of titanium tetrachloride have proved to be commercial methods for producing titanium metal. However, current commercial methods use batch processing which requires significant material handling with resulting opportunities for contamination and gives quality variation from batch to batch. The greatest potential for decreasing production cost is the development of a continuous reduction process with attendant reduction in material handling. There is a strong demand for both the development of a process that enables continuous economical production of titanium metal and for the production of metal powder suitable for use without additional processing for application to powder metallurgy or for vacuum-arc melting to ingot form.
The Kroll process and the Hunter process are the two present day methods of producing titanium commercially. In the Kroll process, titanium tetrachloride is chemically reduced by magnesium at about 1000.degree. C. The process is conducted in a batch fashion in a metal retort with an inert atmosphere, either helium or argon. Magnesium is charged into the vessel and heated to prepare a molten magnesium bath. Liquid titanium tetrachloride at room temperature is dispersed dropwise above the molten magnesium bath. The liquid titanium tetrachloride vaporizes in the gaseous zone above the molten magnesium bath. A reaction occurs on the molten magnesium surface to form titanium and magnesium chloride. The Hunter process is similar to the Kroll process, but uses sodium instead of magnesium to reduce the titanium tetrachloride to titanium metal and produces sodium chloride as a by product.
For both processes, the reaction is uncontrolled and sporadic and promotes the growth of dendritic titanium metal. The titanium fuses into a mass that encapsulates some of the molten magnesium (or sodium) chloride. This fused mass is called titanium sponge. After cooling of the metal retort, the solidified titanium sponge metal is broken up, crushed, purified and then dried in a stream of hot nitrogen. Metal ingots are made by compacting the sponge, welding pieces into an electrode and then melting it into an ingot in a high vacuum arc furnace. High purity ingots require multiple arc melting operations. Powder titanium is usually produced from the sponge through grinding, shot casting or centrifugal processes. A common technique is to first react the titanium with hydrogen to make brittle titanium hydride to facilitate the grinding process. After formation of the powder titanium hydride, the particles are dehydrogenated to produce a usable metal powder product. The processing of the titanium sponge into a usable form is difficult, labor intensive, and increases the product cost by a factor of two to three.
The processes discussed above have several intrinsic problems that contribute heavily to the high cost of titanium production. Batch process production is inherently capital and labor intensive. Titanium sponge requires substantial additional processing to produce titanium in a usable form; thereby increasing cost, increasing hazard to workers and exacerbating batch quality control difficulties. Neither process utilizes the large exothermic energy reaction, requiring substantial energy input for titanium production (approximately 6 kW-hr/kg product metal). In addition, the processes generate significant production wastes that are of environmental concern.